Carotid surgery without arteriography: noninvasive selection of patients.

All carotid noninvasive studies at our institution comprised of duplex scanning, spectral frequency analysis, and ocular-pneumoplethysmography-Gee supraorbital Doppler assessments from 1985-1987 were reviewed. Forty symptomatic and 104 asymptomatic internal carotid arteries, concomitantly studied noninvasively and arteriographically, were identified. All studies were rereviewed prospectively and in blinded fashion. Utilizing peak frequency--internal carotid artery greater than 10 mHz and carotid index (Pf-ICA)/PF-common carotid) greater than 5 as criteria for surgery, 39/40 symptomatic internal carotid arteries were considered appropriate for carotid endarterectomy by noninvasive study. All of these internal carotid arteries had arteriographic confirmation of greater than 50% internal carotid artery stenosis; 22 of them met noninvasive criteria for surgery of peak systolic frequency-internal carotid artery greater than 14 mHz, carotid index greater than 7 and abnormal ocular-pneumoplethysmography-Gee/supraorbital Doppler. All of these had arteriographic confirmation of greater than 80% internal carotid artery stenosis. Eleven asymptomatic internal carotid arteries met spectral frequency criteria for carotid endarterectomy but had normal ocular-pneumoplethysmography-Gee/supraorbital Doppler. Eight in this group had less than 80% stenosis on arteriographic exam. Carotid endarterectomy may be performed without prior arteriography, provided objective criteria are established in a reliable noninvasive lab and met by individual patients.